Inverse Problem Optimization Method to Design Passive Samplers for Volatile Organic Compounds: Principle and Application
| Autor: | Qiujian Xu, Jianping Cao, Yinping Zhang, Zhengjian Du, Jinhan Mo, Xinxiao Li |
|---|---|
| Rok vydání: | 2016 |
| Předmět: |
Air Pollutants
Volatile Organic Compounds 010504 meteorology & atmospheric sciences business.industry Chemistry Sampling (statistics) Sampling error Benzene General Chemistry 010501 environmental sciences Inverse problem 01 natural sciences Benzene measurement Approximation error Mass transfer Formaldehyde Environmental monitoring Electronic engineering Environmental Chemistry Process engineering business 0105 earth and related environmental sciences Passive sampling Environmental Monitoring |
| Zdroj: | Environmental sciencetechnology. 50(24) |
| ISSN: | 1520-5851 |
| Popis: | Passive sampling is an alternative to active sampling for measuring concentrations of gas-phase volatile organic compounds (VOCs). However, the uncertainty or relative error of the measurements have not been minimized due to the limitations of existing design methods. In this paper, we have developed a novel method, the inverse problem optimization method, to address the problems associated with designing accurate passive samplers. The principle is to determine the most appropriate physical properties of the materials, and the optimal geometry of a passive sampler, by minimizing the relative sampling error based on the mass transfer model of VOCs for a passive sampler. As an example application, we used our proposed method to optimize radial passive samplers for the sampling of benzene and formaldehyde in a normal indoor environment. A new passive sampler, which we have called the Tsinghua Passive Diffusive Sampler (THPDS), for indoor benzene measurement was developed according to the optimized results. Silica zeolite was selected as the sorbent for the THPDS. The measured overall uncertainty of THPDS (22% for benzene) is lower than that of most commercially available passive samplers but is quite a bit larger than the modeled uncertainty (4.8% for benzene, the optimized result), suggesting that further research is required. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|